The intent of this paper is to summarize evaluations of AIS made by the bridge teams of three cruise ships.
The test was initiated by the Pacific Marine Review Panel (PACMAR) of the Chamber of Shipping of British Columbia. The main sponsors were Transport Canada, Canadian Coast Guard, Holland America-Line Westours, Inc., Princess Cruises, Northwest Cruise-ship Association and Tideland Signal Corporation.
The test was held during the 4 weeks ending in mid-September, 2000 in British Columbia and South East Alaska. During the whole period, when within AIS' range, SEA PRINCESS and VOLENDAM were able to identify each other, REGAL PRINCESS as well as a small ferry on their electronic chart and on their radar screen. REGAL PRINCESS carried a fully functional transponder the whole time, however, for technical reasons, its bridge team was only able to identify AIS targets on their electronic chart and only during the last two weeks of the test. The installation of a transponder at a base station on Discovery Mountain near Seymour Narrows was not completed until after the test.
Apart from Canadian and Alaskan pilots, the bridge teams consisted mainly of:
- Italian Officers on REGAL PRINCESS (1200 passengers)
- British Officers on SEA PRINCESS (2200 passengers)
- Dutch Officers on VOLENDAM (1200 passengers)
During the test, both REGAL PRINCESS and VOLENDAM made weekly roundtrips to S.E. Alaska from Vancouver through Seymour Narrows. SEA PRINCESS made two fortnightly roundtrips to Seward, Alaska also from Vancouver and also through Seymour Narrows. The ferry POWELL RIVER QUEEN that runs between Campbell River and Quadra Island, crossing the shipping channel just South of Seymour Narrows, carried a fully functional TS/MDS transponder, however, it was not connected to a display.
On SEA PRINCESS and VOLENDAM the transponder from Tideland Signal/Marine Data Systems (TS/MDS) was mounted near the radar transceivers in the base of the foremast. It was connected to the network of the STN Atlas Integrated Bridge System (IBS) through an interface box that was developed by STN Atlas especially for this test and that was installed adjacent to the radar transceivers. The interface allowed display of AIS details and targets on the electronic chart 'Chart Pilot' (see www.uais.org/BCAISTest2000EarlyResults.htm ) and on the main radar screen 'Multi Pilot II' when it was set to Chart Mode (Radar overlay over the electronic chart).
On REGAL PRINCESS the transponder, also from TS/MDS, was mounted on the bridge and connected to Transas Marine's NaviSailor 2500 ECS via a serial cable using interface software that Transas developed especially for this test. REGAL PRINCESS's 'Master' ECS station showed AIS details as well as targets (see VOLENDAM docked in Vancouver below) while the 'Slave' Station showed just targets.
The transponders were set to communicate on VHF Channels 83A and 83B using the Fixed Allocation Time Division Multiple Access (FATDMA) protocol to update positions every 2 seconds, i.e. every 75th time slot. Transponders used an internal, differentially corrected GPS for position references. They used a limited set of proprietary sentences over a 38.4 KB per Second serial cable to communicate with display systems (IBS and ECS). A ship's fixed and voyage data were entered directly into the transponder using a laptop computer and transmitted every 6 minutes. The early version of the transponder used for this test was not yet designed to convey rate of turn information or process other standard AIS sentences (Interrogation, Channel Management, etc.).
Prior to the start of the test, bridge teams were briefed on how AIS works, on the purpose of the test and provided with a list of expected 'AIS encounters' with other AIS equipped ships. For each encounter, they were asked to log first AIS contact, first ARPA contact, last ARPA contact and last AIS contact. They were also asked to evaluate AIS as an aid to navigation using a an open ended set of criteria (see http://www.uais.org/InstructionstoMasterSEAPRINCESS.htm) and to present their evaluation during the debriefing which took place at the end of the test.
Following is a compilation of the minutes of debriefing meetings held in Vancouver on September 9 on SEA PRINCESS, on September 10 on REGAL PRINCESS and on September 11 on VOLENDAM. Also included are comments from the AIS Evaluation document REGAL PRINCESS distributed during the debriefing. The latter are in quotation marks. Evaluation issues discussed are organized by the evaluation criteria that Bridge Teams were asked to use.
- SEA PRINCESS
Felt that AIS should take care of most if not all current reporting requirements to a VTS center.
Expressed that AIS might give the OOW a false sense of security if no AIS targets showed in radar blind spots. We discussed ways to increase the reliability of AIS reception through repeater stations at crucial locations and through transmitting AIS (back-up) information via satellite.
- REGAL PRINCESS
"Our general comment about the use and implementation of AIS system it is certainly quite short, as well as it was the time we had to evaluate its functionality. The system it is not properly installed yet, as we did not see any target in Discovery Mountain base station. We have not seen yet any AIS target at all displayed on the radar screen. Though, the time since we had it partially installed it is too short for an effective evaluation. If we would have the Transponder installed since the beginning of the Alaskan cruise (season), surely that by now we would all know more about its pro’s and con’s."
"Also, as we often are monitoring both radar and the electronic chart on a quite short range, due to the narrow passages of Alaskan cruises, it will be unluckily that we will ever be able to pay the proper attention on the range, which AIS could detect a target equipped with the same system. Often, it is a good rule to pay as much attention as possible to what it is really happening in the vicinity of the ship, more than to extend the range looking for potential other targets. We are quite confident that, once the system will be properly evaluated and improved by the professional technicians, with the help of user(s), it will certainly be more effective and useful than it looked so far."
"We believe that it is not so useful to fill the Radar screen with an excess of info about targets. Unless there is the option to switch on/off the AIS info on the radar, it could be distracting to read too many things, especially whilst navigating in very congested areas. On the other hand, it is certainly useful to have more information about targets behind blind point, like Discovery passage or Johnstone passage, on the electronic chart."
Because radar does not always pick up log booms, it would be very beneficial to equip them or their tugs with a transponder.
The bridge team felt that ships and boats that are equipped with a transponder may start to rely on it too much. They may feel that because they have a transponder, large ships would definitely see them and stay out of their way, regardless of rules of the road.
It was felt that everything possible should be done to allow the OOW to trust that AIS targets in radar blind spots will show. This includes using shore based repeater stations near radar blind spots and using satellite rather than VHF communication to increase the reliability of communication between transponders.
- VOLENDAM
Peter Bouwmeester, 3rd Officer & Navigator, noted that sometimes ARPA targets wander away and fuse with the shore in narrow channels. He felt that AIS would be particularly helpful to avoid these occurrences.
The general consensus was that AIS is extremely useful for identification of other ships in the area, but that its implementation should not cause information overflow for the OOW.
- SEA PRINCESS
Display of AIS targets of non-SOLAS ships smaller than 20 m LoA should be user selectable. In heavy traffic situations their display will need to turned off.
- REGAL PRINCESS
"We do not think this is a useful option. Often, especially on departure/arrival in busy ports like those where cruise ships use to call, we deal with a large number of pleasure boats and/or fishing boats. For the sake of a good anti collision watch, we cannot plot every target displayed, as it will just confuse the observer even more, with the consequence of not focusing properly on the dangerous targets, with short CPA. It could be eventually considered a helpful option if the above-mentioned switch facility would be available on the Radar."
- VOLENDAM
The team felt that showing all AIS equipped targets would be helpful but wanted to have the option to turn off display of targets that represent ships smaller than 20 m LoA.
- REGAL PRINCESS
"The Solas target minimum range would be 24 mls. The non Solas targets 12 mls. "
"The minimum TCPA would be 30 minutes."
AIS information will only be needed for targets that are in a forward looking guard zone.
- SEA PRINCESS
BCCP pilots were generally more interested in the AIS test because they had heard about it through BCCP and PPA. Some SEAPA pilots also showed an interest, particularly Messrs. Winters, Kellogg and Clinkshales.
- REGAL PRINCESS
"We have tried to discuss the use and the features of AIS with BC Pilots and AK Pilots. Most of them are not even aware of this system; those who know about it have different opinion. For example, we know that BC Pilots are generally not very happy about having displayed too much info on the radar. That often leads to confusion and takes away the attention from dangerous situations. They carried out test over about 10 ships in Georgia strait, and the result was quite confusing with quite a congested series of not essential information displayed."
"We believe that, as a general rule, Pilots thinks that the increasing amount of hi-tech utilities, especially during the passage of dangerous area like Seymour Narrows, are taking away the concentration of Officers from the ‘real thing’, the sight navigation and ARPA monitoring."
- SEA PRINCESS
SEA PRINCESS's transponder worked great in the beginning but later on the range at which AIS targets were received became short (< 12 Nm). We inferred that if other ships were able to 'see' SEA PRINCESS from a greater range (this turned out to be true) that then the receiver of SEA PRINCESS's transponder must have a problem. We will ask REGAL PRINCESS and VOLENDAM about this and attempt to ship a replacement transponder to Juneau.
The lack of diagnostic capabilities was noted as a shortcoming by Steve Pegg, Sr. Communications Officer. He suggested that there should be an easy way to not only check the performance of the receivers and the transmitter, but also an easy way to check the static and voyage data that are being broadcast by the transponder. An alarm should possibly be set up to warn the OOW on the bridge that something is awry with transponder performance.
Mr. Pegg, also suggested that transponders should be surveyed periodically just like the GMDSS equipment.
- REGAL PRINCESS
"We cannot list any technical aspect, i.e. maintenance or operation, since it is too short the evaluation period."
It was felt that, at least on SOLAS ships, a back-up transponder should be on standby. The OOW should be able to switch to the back-up as soon as an alarm indicates that the primary transponder is not performing properly. Switching to the back-up transponder should be made as easy as possible. We did not discuss automatic fail-over to the back-up transponder, however I believe that the bridge team would prefer automatic over manual fail-over.
- REGAL PRINCESS
The bridge team felt that radar is the main navigation instrument and that AIS information should be displayed on it as needed.
It was generally felt that AIS should function as much as possible like ARPA because everybody is used to using ARPA information.
- VOLENDAM
AIS targets were felt to be trailing behind ARPA targets of the same ship. The difference in the bearing to an ARPA target and an AIS target was noted to be as much a 3 degrees.
The team agreed that it was important that AIS, its capabilities and its limitations be incorporated in the curriculum of maritime colleges.
- SEA PRINCESS
The bridge team felt that AIS targets should show their predicted path as a curved line rather than show ROT separate from a vector for the target's COG and SOG.
In potential collision situations the bridge team uses relative rather than true values for assessing the traffic situation because doing so eliminates currents as a source of distortion of the situation.
- REGAL PRINCESS
We discussed the display alternatives for an AIS target's Course Over Ground, Speed Over Ground and Rate of Turn. The consensus was that showing them all three combined in the form of the predicted path would be the most intuitive.
- VOLENDAM
The team felt that if an AIS targets shows a vector it should be a curved one representing the predicted path, i.e. take into account the current course, current speed and current ROT.
- SEA PRINCESS
Use (abbreviation of) ship name rather than Callsign, MMSI number or IMO number to label AIS targets.
The interval of 6 minutes between receiving updates of the ship name (and other static and voyage data) is too long. This interval should, like position updates be shorter for faster ships.
There was a question about what number was used to label an AIS target before it was associated with a name (it seemed to be a sequential number).
The label of an AIS target has the same font and color as other place names on the ECDIS screen. There should be a clear difference.
- REGAL PRINCESS
1st Priority: "Label targets with ship’s name and/or call sign"
- VOLENDAM
The team felt that AIS targets, when 'plotted' (clicked-on) should provide the same information as ARPA targets do (CPA, TCPA, Course, Speed, Bearing and Range). It was not felt that it was necessary to label AIS targets with (an abbreviation of) the name of the ship. Most of the team felt that displaying an AIS icon as a silhouette, showing its length and breadth to scale, would be helpful in close quarters. The team felt that it wouldn't be necessary to show a vector emanating from an AIS target if the target was not of immediate concern, especially for targets that represent ships smaller than 20 m.
- SEA PRINCESS
We discussed target consolidation and ranking of ARPA versus AIS target information. This discussion left me with the impression that AIS information will have to gain your trust before its value can be assessed. The bridge team felt that consolidation is required and that rules are required to automatically do so based on tight criteria for (trend in) distance between targets, (trend in) difference between relative ARPA and AIS course and (trend in) difference between relative AIS and ARPA speed. Whether AIS information is ranked more reliable that ARPA information of the same target should also depend on the quality (HDOP or Estimated Position Error) of the dGPS signal that the AIS target receives.
- VOLENDAM
The team agreed that icons that represent the same ship should be consolidated. Until the team gained experience and confidence in the accuracy of AIS information, it preferred to use radar information to describe the course, speed and position of the consolidated icon (Better the devil you know).
- SEA PRINCESS
The bridge team felt that it would be useful to receive re-broadcasts of ARPA targets that are acquired by VTS centers, if they are consolidated with the ARPA targets for the same ship that are received by the ships own radar. It would be even more useful if these had (abbreviations) of ship names associated with them.
- REGAL PRINCESS
2nd Priority: "Relay ARPA target info to others"
- VOLENDAM
The team agreed that only VTS centers should be allowed to re-broadcast AIS information or broadcast ARPA targets as pseudo AIS targets and then only if target consolidation works well.
- VOLENDAM
As far as using traffic related e-mail between ships, the team felt that one of the advantages of VHF is that you can listen in on a conversation between other ships in the area or between the local VTS center and other ships. E-mail, if used, should emulate this feature by making it work like a bulletin board. Most team members, however, felt that watching such a bulletin board would be distracting to the OOW.
- VOLENDAM
The team didn't feel that hazardous cargo information was very useful for bridge operations.
- SEA PRINCESS
AIS targets should allow display of comments entered into the system by the OOW of an AIS target to explain, for instance, maneuvering restrictions.
- REGAL PRINCESS
3rd Priority: "Display info about target unusual maneuvering limitations"
- VOLENDAM
It was felt that if an AIS equipped ship has maneuvering restrictions, that then the OOW of that ship should be able to enter more details about the nature of the restrictions, i.e. 'SB prop inoperable'. This information should be available along with other details (GRT, Ship Type, LoA, Draft, number of barges in tow, destination, ETA, etc.) in a window that is opened by clicking on a 'more...' button in the primary target information window. This secondary window could be displayed on a separate screen.
- SEA PRINCESS
AIS transponders should automatically change their status from 'at anchor' or 'at dock' to 'underway' if its SOG is no longer zero or near zero (averaged over a short period). This should not depend on the OOW changing the status.
- VOLENDAM
It was deemed helpful if the ship type could be displayed by using a different color icon for a tug, a fishing boat, a pilot boat, etc.
- SEA PRINCESS
The bridge team thought it would be useful to have Nav Aids transmit their status and be able to selectively and temporarily show this status. We discussed 'virtual' bouys that could replace actual bouys, but the bridge team didn't feel comfortable with that idea. Showing traffic separation zones on the screen was seen as useful. Their source could be the electronic chart or AIS-like transmissions from a base stations.
It was felt that receiving actually observed current velocities at Seymour Narrows would be extremely useful.
- VOLENDAM
The team felt that it might be useful to have a displaced bouy show its actual position through AIS and to allow identification of lights through AIS as a reference to what is observed visually. The team also felt that changes in the status or new navigational aids should be displayed via AIS prior to being listed in 'Notices to Mariners'.
The team expressed interest in showing the actual velocity of the current in Seymour Narrows, but emphasized that it would be more useful to display actual observation of other environmental conditions (visibility, wind, sea-state, etc.) that are used in deciding which channel/route to sail.
- SEA PRINCESS
You felt that it would be useful if there was an option to, selectively and temporarily, display and AIS target's intended route.
Tracy Fleming, 3rd Officer, suggested that it should be possible to ask ECDIS to list all AIS targets that have an ETA at a particular (way) point similar to your own ETA there, so that you know which ships to contact to discuss passing arrangements.
- REGAL PRINCESS
4th Priority: "Display target voyage info"
The bridge team felt it would be helpful to be able to find out from AIS which ships have an ETA at a narrow channel (like Seymour Narrows) similar to their own and what that ETA is. The team noted that currently ships sometimes report (via VHF) a very optimistic ETA at Seymour Narrows. The team felt that this was done to ensure earlier passage through the narrows. AIS is expected to temper such optimism. Currently, ships approaching Seymour Narrows from the opposite direction are unable to verify the reported ETA because their radar can not acquire the approaching vessel.
- VOLENDAM
The team felt that it would be useful if AIS could display a list of names of AIS equipped ships that have an ETA at a particular waypoint similar to the own ship's ETA there, to allow contacting them about passing arrangements and to select the safest route/channel to sail.
- SEA PRINCESS
It was felt that there might be some value in showing the trustworthiness of a target's position is based on the quality (HDOP or Estimated Position Error) of the dGPS signal that the AIS target receives and the distance between it and the ARPA target for the same ship.
- REGAL PRINCESS
It was felt that the accuracy of the position of an AIS target should be available to the OOW. Accuracy should be expressed in Estimated Position Error (EPE) or HDOP rather than in terms of 'High' or 'Low', as is currently planned.
- VOLENDAM
The team felt that it would be very helpful if each OOW could have his preferences for display options saved and easily reactivated when starting a new watch.
Peter Bouwmeester, 3rd Officer & Navigator, showed examples of a 'stray' AIS target with an MMSI number for 'Noordster' YE 158, a Dutch fishing boat, just off the Sisters Islands in the Strait of Georgia.
He also showed an instance where REGAL PRINCESS was represented by two separate icons, one with the correct MMSI number and one with an incorrect number.
- Display a target's details in a separate window upon "mouse-over"
- Display a target's details in a separate window upon "mouse-click"
- REGAL PRINCESS
"Display a target’s details in a separate window upon a mouse click"
- Display details of all targets in a table sorted by their TCPA on a separate screen page.
- Other (Specify)
- SEA PRINCESS
The bridge team felt that the AIS Info window was too large. It hides too much of the screen and might interfere with safe navigation. It was felt that, like ARPA information, AIS information should appear in a window off the active part of the screen when an AIS target is selected with the cursor. It was, also, felt that this window should not just provide a snapshot of AIS information, but that it should always show the latest available information. Furthermore it was felt that it would be useful to enlarge an AIS target's name label to show CPA, TCPA, COG, SOG, bearing and range when the target is selected with a mouse button. This information should disappear upon release of the mouse button.
- VOLENDAM
The team had a strong preference for displaying AIS information: off the main screen and to the side, just like ARPA target information is currently displayed. There should however be a clear distinction between a window showing details of an ARPA target and a window showing AIS details. The data in the AIS window needs to stay up-to-date and not be a single snapshot as is now the case.
- Close a target's details window after 30 seconds
- Close a target's details window upon "mouse-off"
- Close a target's details window upon "mouse-click"
- REGAL PRINCESS
"Close a target’s details window upon a mouse click"
- Other (Specify)
